Kevin P. Oh

Deep learning models that predict molecular phenotypes directly from DNA sequence offer a powerful framework for interpreting genomic variation. Recently, AlphaGenome was introduced as a deep sequence-to-function architecture capable of predicting observations that historically required experiments. While the model has shown high accuracy, it was primarily evaluated on human variants scored against a reference genome. Here, we test performance on mouse data, the other species AlphaGenome was trained on although with fivefold fewer features than human (1,128 versus 5,930). We demonstrate that AlphaGenome’s predictive performance varies considerably depending on the functional task. Specifically, predicted quantitative expression effects are directionally weak and compressed roughly 100-fold relative to empirical benchmarks across both reconstructed-haplotype and single-variant regimes. In contrast, canonical splice-site disruptions are recognized with near-identical accuracy in mouse and human (AUC 0.96 versus 0.98), displaying no cross-species divergence in predicted effect magnitude. We developed a scoring-approach for AI-agents to autonomously assess AlphaGenome prediction confidence and accurately differentiate between AlphaGenome’s robust sequence-level recognition across species and its current limitations when interpreting un-fine-mapped regulatory variants. This demonstrates how GenAI innovations that are still under development can safely be harnessed by wrapping a responsible AI layer around the call to intercept flawed results, thereby adhering to international standards, such as the Australian Voluntary AI Safety Standard (VAISS).

The size and distribution of home ranges reflect how individuals within a population use, defend, and share space and resources, and may thus be an important predictor of population‐level dynamics. Eruptive species, such as the house mouse in Australian grain‐growing regions, are an ideal species in which to investigate variations in space use and home range overlap between stable and outbreaking populations. In this study, we use spatially explicit capture–recapture models to explore if space use and home range overlap among female mice could serve as indicators of changes in population density leading into summer. Additionally, we assess the sensitivity of space use and home range estimates to reduced recapture rates. Our analysis did not reveal variations in the spring spatial organisation of female mice based on existing capture–mark–recapture data. However, our study highlights the need to balance monitoring efforts within regions, emphasising the importance of exploring studies that can improve spatial recaptures by optimising trapping efforts. This is particularly important in Australian agricultural systems, where varying farm management practices may drive differences in population dynamics.

The management of invasive species has been greatly enhanced by population genetic analyses of multilocus single-nucleotide polymorphism (SNP) datasets that provide critical information regarding pest population structure, invasion pathways, and reproductive biology. For many applications there is a need for protocols that offer rapid, robust and efficient genotyping on the order of hundreds to thousands of SNPs, that can be tailored to specific study populations and that are scalable for long-term monitoring schemes. Despite its status as a model laboratory species, there are few existing resources for studying wild populations of house mice (Mus musculus spp.) that strike this balance between data density and laboratory efficiency. Here we evaluate the utility of a custom targeted capture genotyping-by-sequencing approach to support research on plaguing house mouse populations in Australia. This approach utilizes 3,651 hybridization capture probes targeting genome-wide SNPs identified from a sample of mice collected in grain-producing regions of southeastern Australia genotyped using a commercially available microarray platform. To assess performance of the custom panel, we genotyped wild caught mice (N = 320) from two adjoining farms and demonstrate the ability to correctly assign individuals to source populations with high confidence (mean >95%), as well as robust kinship inference within sites. We discuss these results in the context of proposed applications for future genetic monitoring of house mice in Australia.

Invasive rodents are a major cause of environmental damage and biodiversity loss, particularly on islands. Unlike insects, genetic biocontrol strategies including population-suppressing gene drives with biased inheritance have not been developed in mice. Here, we demonstrate a gene drive strategy ( t _CRISPR ) that leverages super-Mendelian transmission of the t haplotype to spread inactivating mutations in a haplosufficient female fertility gene ( Prl ). Using spatially explicit individual-based in silico modeling, we show that t _CRISPR can eradicate island populations under a range of realistic field-based parameter values. We also engineer transgenic t _CRISPR mice that, crucially, exhibit biased transmission of the modified t haplotype and Prl mutations at levels our modeling predicts would be sufficient for eradication. This is an example of a feasible gene drive system for invasive alien rodent population control.

Invasive rodents, including house mice, are a major cause of environmental damage and biodiversity loss, particularly in island ecosystems. Eradication can be achieved through the distribution of rodenticide, but this approach is expensive to apply at scale, can have negative impacts (e.g. on non-target species, or through contamination), has animal ethics concerns, and has restrictions on where it can be used. Gene drives, which exhibit biased inheritance, have been proposed as a next generation strategy to control invasive alien pests and disease vectors. However, synthetic gene drives including CRISPR homing drives have proven to be technically challenging to develop in mice. The t haplotype is a naturally-occurring segregation distortion locus with highly biased transmission from heterozygous males. Here we propose a novel gene drive strategy for population suppression, t _CRISPR , that leverages t haplotype bias and an embedded SpCas9/gRNA transgene to spread inactivating mutations in a haplosufficient female fertility gene. Using spatially explicit individual-based in silico modelling, we show that polyandry, sperm competition, dispersal, and transmission bias are critical factors for t _CRISPR -mediated population suppression. Modelling of realistic parameter values indicates that t _CRISPR can eradicate an island population of 200,000 mice while the unmodified t haplotype fails under the same conditions. We also demonstrate feasibility of this approach by engineering t _CRISPR mice in a safe split drive format. t _CRISPR mice exhibit high transmission of the modified t haplotype, and efficient generation and transmission of inactivating mutations in a recessive female fertility gene, crucially, at levels for which the modelling predicts that population eradication can occur. This is the first example of a feasible gene drive system for invasive alien rodent population control.

Sexual signalling traits are often observed to diverge rapidly among populations, thereby playing a potentially key early role in the evolution of reproductive isolation. While often assumed to reflect divergent sexual selection among populations, patterns of sexual trait diversification might sometimes be biased along axes of standing additive genetic variation and covariation among trait components. Additionally, theory predicts that environmentally induced phenotypic variation might facilitate rapid trait evolution, suggesting that patterns of divergence between populations should mirror phenotypic plasticity within populations. Here, we evaluate the concordance between observed axes of multivariate sexual trait divergence and predicted divergence based on (1) interpopulation variation in sexual selection, (2) additive genetic variances and (3) temperature‐related phenotypic plasticity in male courtship song among geographically isolated populations of the Hawaiian swordtail cricket, Laupala cerasina, which exhibit sexual isolation due acoustic signalling traits. The major axis of multivariate divergence, d_max, accounted for 76% of variation among population male song trait means and was moderately correlated with interpopulation differences in directional sexual selection based on female preferences. However, the majority of additive genetic variance was largely oriented away from the direction of divergence, suggesting that standing genetic variation may not play a dominant role in the patterning of signal divergence. In contrast, the axis of phenotypic plasticity strongly mirrored patterns of interpopulation phenotypic divergence, which is consistent with a role for temperature‐related plasticity in facilitating instead of inhibiting male song evolution and sexual isolation in these incipient species. We propose potential mechanisms by which sexual selection might interact with phenotypic plasticity to facilitate the rapid acoustic diversification observed in this species and clade.

Introduced rodent populations pose significant threats worldwide, with particularly severe impacts on islands. Advancements in genome editing have motivated interest in synthetic gene drives that could potentially provide efficient and localized suppression of invasive rodent populations. Application of such technologies will require rigorous population genomic surveys to evaluate population connectivity, taxonomic identification, and to inform design of gene drive localization mechanisms. One proposed approach leverages the predicted shifts in genetic variation that accompany island colonization, wherein founder effects, genetic drift, and island‐specific selection are expected to result in locally fixed alleles (LFA) that are variable in neighboring nontarget populations. Engineering of guide RNAs that target LFA may thus yield gene drives that spread within invasive island populations, but would have limited impacts on nontarget populations in the event of an escape. Here we used pooled whole‐genome sequencing of invasive mouse (Mus musculus) populations on four islands along with paired putative source populations to test genetic predictions of island colonization and characterize locally fixed Cas9 genomic targets. Patterns of variation across the genome reflected marked reductions in allelic diversity in island populations and moderate to high degrees of differentiation from nearby source populations despite relatively recent colonization. Locally fixed Cas9 sites in female fertility genes were observed in all island populations, including a small number with multiplexing potential. In practice, rigorous sampling of presumptive LFA will be essential to fully assess risk of resistance alleles. These results should serve to guide development of improved, spatially limited gene drive design in future applications.

Invasive species are increasingly affecting agriculture, food, fisheries, and forestry resources throughout the world. As a result of global trade, invasive species are often introduced into new environments where they become established and cause harm to human health, agriculture, and the environment. Prevention of new introductions is a high priority for addressing the harm caused by invasive species, but unfortunately efforts to prevent new introductions do not address the economic harm that is presently manifested where invasive species have already become established. Genetic biocontrol can be defined as the release of organisms with genetic methods designed to disrupt the reproduction of invasive populations. While these methods offer the potential to control or even eradicate invasive species, there is a need to ensure that genetic biocontrol methods can be deployed in a way that minimizes potential harm to the environment. This review provides an overview of the state of genetic biocontrol, focusing on several approaches that were the subject of presentations at the Genetic Biocontrol for Invasive Species Workshop in Tarragona, Spain, March 31st, 2019, a workshop sponsored by the OECD’s Co-operative Research Program on Biological Resource Management for Sustainable Agricultural Systems. The review considers four different approaches to genetic biocontrol for invasive species; sterile-release, YY Males, Trojan Female Technique, and gene drive. The different approaches will be compared with respect to the efficiency each affords as a genetic biocontrol tool, the practical utility and cost/benefits associated with implementation of the approach, and the regulatory considerations that will need to be addressed for each. The opinions expressed and arguments employed in this publication are the sole responsibility of the authors and do not necessarily reflect those of the OECD or of the governments of its Member countries.

Invasive rodents impact biodiversity, human health and food security worldwide. The biodiversity impacts are particularly significant on islands, which are the primary sites of vertebrate extinctions and where we are reaching the limits of current control technologies. Gene drives may represent an effective approach to this challenge, but knowledge gaps remain in a number of areas. This paper is focused on what is currently known about natural and developing synthetic gene drive systems in mice, some key areas where key knowledge gaps exist, findings in a variety of disciplines relevant to those gaps and a brief consideration of how engagement at the regulatory, stakeholder and community levels can accompany and contribute to this effort. Our primary species focus is the house mouse, Mus musculus , as a genetic model system that is also an important invasive pest. Our primary application focus is the development of gene drive systems intended to reduce reproduction and potentially eliminate invasive rodents from islands. Gene drive technologies in rodents have the potential to produce significant benefits for biodiversity conservation, human health and food security. A broad-based, multidisciplinary approach is necessary to assess this potential in a transparent, effective and responsible manner.